Reclassification & Cost Surface

Four Reclassified Layers

Methodology: Each variable was reclassified to a standardized 1-5 scale using natural breaks classification, enabling direct comparison across different metrics.

Python Integration: VIIRS, 311, Crime, and Census data were first integrated into philly_integrated_data.shp using geopandas

ArcGIS Processing: Reclassify tool applied to each variable → Raster Calculator combined layers

Layer 1: Darkness (VIIRS Inverted)

Make this Notebook Trusted to load map: File -> Trust Notebook

Darkness Score: Higher values = darker areas (inverted VIIRS radiance)

Range: 0.000000 - 1.000000 reclassified to 1-5 scale

Pattern: Center City shows lowest darkness scores; North and Southwest Philadelphia show highest

Layer 2: Low 311 Reporting

Make this Notebook Trusted to load map: File -> Trust Notebook

Low Reporting Score: Higher values = fewer complaints filed

Range: 0.000000 - 1.000000 reclassified to 1-5 scale

Insight: Areas with high darkness but low complaints indicate reporting gaps

Layer 3: Low Income / Vulnerability

Make this Notebook Trusted to load map: File -> Trust Notebook

Vulnerability Score: Based on below-median household income

Range: 0.000000 - 1.000000 reclassified to 1-5 scale

Pattern: North Philadelphia and Southwest show highest vulnerability

Layer 4: Crime Risk

Make this Notebook Trusted to load map: File -> Trust Notebook

Crime Score: Nighttime incident density (7pm-6am)

Range: 0.000000 - 1.000000 reclassified to 1-5 scale

Correlation: High crime areas often overlap with low lighting zones

Weighted Cost Surface

Layer 5: Integrated Risk Surface

Make this Notebook Trusted to load map: File -> Trust Notebook

Cost Surface Formula

Raster Calculator:

CostSurf_Risk = 
    0.30 × Darkness +
    0.30 × Crime +
    0.20 × Low311 +
    0.20 × LowIncome

Interpretation: Brighter colors = higher risk scores = priority intervention areas

Pattern: Risk concentrates in North Philadelphia, Kensington, and Southwest corridors

Neighborhood Analysis: VIIRS Focal Mean

Key Finding

Center City is brightest; peripheral neighborhoods show significantly lower lighting levels

Focal Statistics smoothed VIIRS radiance using circular neighborhood. This revealed that downtown concentration of brightness creates stark inequality with surrounding residential areas.

Complaint Gap Areas

62 Underserved Neighborhoods Identified

Make this Notebook Trusted to load map: File -> Trust Notebook
62 Gap Areas Identified

Where darkness, low complaints, and low income overlap

$42,800 Average Income

38% below city average

0.31 Average VIIRS

65% darker than average

2.1 Complaints/km²

76% fewer than average

Gap Area Characteristics

Three criteria overlap: 1. Low brightness (VIIRS below threshold) 2. Low complaints (311 below city average) 3. Low income (Census below median)

Identified with: Python spatial analysis using geopandas on Zonal_VIIRS_Mean output

Critical Pattern: The darkest areas file 76% fewer complaints. These are also the lowest-income neighborhoods. This means the 311 system systematically misses the communities that need help most.

582 Block Groups Predicted to Continue Darkening

Make this Notebook Trusted to load map: File -> Trust Notebook

582 Areas Predicted to Darken

80% Currently Darkening

Of Philadelphia block groups showed decreasing light 2022-2024

R² = 0.612 Model Performance

Random Forest Regressor

Model Configuration

Algorithm: Random Forest Regressor (scikit-learn)

Hyperparameter Tuning: GridSearchCV for optimal parameters

Top Predictive Features: - Historical brightness trend (-0.035 per year average) - Current VIIRS level - 311 complaint density - Income & population

Validation: Cross-validation R² = 0.612

Urgent Finding

Philadelphia lighting is declining citywide.

Model predicts 582 areas will continue to darken. 80% of Philadelphia already showing decline. Priority intervention needed before infrastructure failure.

Streetlights date to the 1970s. Our 2022-2024 data captures the critical transition before the city’s $91M LED replacement project (launched Aug 2023).

Safe Path Analysis

Network-Based Routing with Risk Weighting

Make this Notebook Trusted to load map: File -> Trust Notebook

Route Analysis Results

52 routes analyzed between key Philadelphia locations: - City Hall - Temple University
- 30th Street Station - University City - Chinatown - Major transit hubs

Methods: - ArcGIS: Distance Accumulation + Optimal Path - Python: OSMnx + NetworkX for street-level routing with risk-weighted edges

TRADE-OFF
+1-3% Distance Increase
=

Up to 20% Safer Routes

Key Finding

Chinatown route shows greatest risk reduction

By accepting a small distance penalty (1-3% longer), travelers can reduce their exposure to high-risk areas by up to 20%.

Priority Areas for Intervention

Make this Notebook Trusted to load map: File -> Trust Notebook

Priority Scoring Methodology

Priority areas were identified by combining: 1. Complaint Gap status (weight: 40%) 2. Predicted darkening (weight: 30%) 3. Current risk score (weight: 20%) 4. Population density (weight: 10%)

Result: Ranked list of neighborhoods requiring immediate intervention

Technical Implementation Notes

Workflow Summary

  1. Python (geopandas): Integrated VIIRS, 311, Crime, Census philly_integrated_data.shp

  2. ArcGIS Reclassify: Standardized each variable to 1-5 scale

  3. ArcGIS Raster Calculator: Combined layers CostSurf_Risk

  4. ArcGIS Spatial Analysis: Zonal Statistics + Focal Statistics

  5. Python Analysis: Identified Complaint Gap areas using spatial overlay

  6. scikit-learn: Trained Random Forest for darkening prediction

  7. OSMnx + NetworkX: Generated risk-weighted safe routes

Next Steps

Continue to Findings to see policy recommendations and conclusions.